Chinese dialect tone’s recognition using gated spiking neural P systems

Tone is the changing trend of pitch with time. In Chinese, tone plays an essential role for distinguishing meaning. Chinese dialect’s tone is more complex with Mandarin. In the field of Chinese dialect phonetics research, using human earing to recognize the types of tones is still the main method. So batch processing is not possible. In this paper, we construct a GSNP (gated spiking neural P) model with 2 layers which can process time series data to recognize the tones of Chinese dialects. The average accuracy rate of seven cities’ speech is more than 97%. Even in the case of small training samples, compared with other methods, the GSNP model has simpler structure, higher accuracy and more efficiency. It can not only improve the work efficiency of Chinese dialect field investigation, but also help researchers to screen the sounds with special sounds.